Upload
duongcong
View
219
Download
1
Embed Size (px)
Citation preview
TitleGOLD NANOPARTICLES-FUNCTIONALIZED GRAPHENE OXIDENANORIBBONS FOR ENHANCED PERFORMANCE OFELECTROCHEMICAL BASED-BIOSENSORS
Author(s) Ismail, Nur Syakimah Binti
Citation
Issue Date
Text Version ETD
URL https://doi.org/10.18910/52141
DOI 10.18910/52141
rights
Doctoral Dissertation
GOLD NANOPARTICLES-FUNCTIONALIZED GRAPHENE OXIDE NANORIBBONS FOR ENHANCED PERFORMANCE OF
ELECTROCHEMICAL BASED-BIOSENSORS
Nur Syakimah Ismail
January 2015
Graduate School of Engineering Osaka University
ii
GOLD NANOPARTICLES-FUNCTIONALIZED GRAPHENE OXIDE
NANORIBBONS FOR ENHANCED PERFORMANCE OF
ELECTROCHEMICAL BASED-BIOSENSORS
(
)
A dissertation submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
by
Nur Syakimah Ismail
Department of Precision Science & Technology and Applied Physics
Graduate School of Engineering
Osaka University
iii
TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION
1.1 Biosensors..1
1.1.1 Electrochemical biosensors....2
1.1.2 Electrochemiluminescence biosensors...3
1.1.3 Nanomaterial-based electrochemical biosensors........6
1.2 Gold nanoparticle in electrochemical biosensor....7
1.3 Graphene for electrochemical applications......12
1.3.1 Graphene......12
1.3.2 Electrochemical performance of graphene ..14
1.3.3 Graphene oxide nanoribbons16
1.4 Organization of thesis...20
1.5 References20
CHAPTER 2: EFFECT OF MWCNT DIAMETER AND CHEMICAL OXIDATION
TIME ON FORMATION AND PERFORMANCE OF GRAPHENE OXIDE
NANORIBBONS FOR ELECTROCHEMICAL BIOSENSOR APPLICATIONS
2.1 Introduction..25
2.2 Experimental Procedure ......................................28
2.2.1 Materials and reagents .28
2.2.2 GONR synthesis and characterization .28
2.2.3 GONRs in biosensor applications ...29
2.3 Results and Discussion 30
2.3.1 Morphology inspection.30
2.3.2 FTIR characterization...33
iv
2.3.3 Raman analysis.37
2.3.4 Investigation on the degree of conjugation..39
2.3.5 Electrochemical performance and sensor applications41
2.4 Conclusions 48
2.5 References....49
CHAPTER 3: DEVELOPMENT OF NON-ENZYMATIC ELECTROCHEMICAL
GLUCOSE SENSOR BASED ON GRAPHENE OXIDE NANORIBBONS-GOLD
NANOPARTICLE HYBRID
3.1 Introduction..53
3.2 Experimental .......................................................56
3.2.1 Materials and reagents .56
3.2.2 Synthesis of GONRs ...56
3.2.3 Synthesis of AuNPs .........57
3.2.4 Fabrication of AuNPs/GONRs/CS electrode...57
3.2.5 Characterization of GONR materials and AuNPs/GONRs/CS electrodes...58
3.3 Results and Discussion 58
3.3.1 Preparation and morphological characterization of AuNP/GONR/CS....58
3.3.2 Characterization of GONRs ........60
3.3.3 Glucose oxidation reaction in neutral conditions ....62
3.3.4 Optimization of GONR and AuNP loadings ..67
3.3.5 Sensitivity and reproducibility 70
3.3.6 Selectivity.74
3.4 Conclusions .79
3.5 References....80
CHAPTER 4: ENHANCED ELECTROCHEMILUMINESCENCE OF N-
v
(AMINOBUTYL)-N-(ETHYLISOLUMINOL) FUNCTIONALIZED GOLD
NANOPARTICLES BY GRAPHENE OXIDE NANORIBBONS
4.1 Introduction..83
4.2 Experimental .......................................................86
4.2.1 Materials and reagents .86
4.2.2 Synthesis of GONRs ...86
4.2.3 Synthesis of ABEI-AuNPs ..............87
4.2.4 Fabrication of ABEI-AuNP and ABEI-AuNP-GONR modified SPE ....87
4.2.5 Characterization of ABEI-AuNP-GONR material and ECL measurement.88
4.3 Results and Discussion 89
4.3.1 Characterization of ABEI-AuNP-GONR/SPE ....89
4.3.2 ECL of ABEI-AuNP-GONR modified SPE ...............91
4.3.3 Effect of loading ratio on ECL of ABEI-AuNP-GONR/SPE .......101
4.3.4 Effect of pH, H2O2 and O2 on ECL of ABEI-AuNP-GONR/SPE.104
4.3.5 Propose ECL mechanism of ABEI-AuNP-GONR/SPE 107
4.3.6 Characterization interaction between ABEI-AuNP and GONR ...108
4.3.7 Application in biosensors ..109
4.4 Conclusions ...112
4.5 References......113
CHAPTER 5: SUMMARY & FUTURE REMARKS
5.1 Summary.....117
5.2 Future Remarks..121
LIST OF PUBLICATIONS.....122
PRESENTATIONS AT SCIENTIFIC MEETING....123
ACKNOWLEDGEMENTS.....125
vi
LIST OF ABBREVIATIONS
3-AP 3-amino-phthalate
3-AP* Excited state of 3-amino-phthalate
AA Ascorbic acid
ABEI N-(aminobutyl)-N-(ethylisoluminol)
ABEI-AuNP Gold Nanoparticle coated ABEI molecules
Ag/AgCl Silver/Silver Chloride reference electrode
AP -acetamidophenol
ATR-FTIR Attenuated Total Reflection Fourier Transform Infrared
AuNP Gold Nanoparticle
CA Chronoamperometry
CBS Carbonate Buffer Solution
CE Counter Electrode
CL Chemiluminescence
CNT Carbon Nanotube
CS Carbon sheet
CV Cyclic Voltammetry
ECL Electrochemiluminescence
EPPG Edge-Plane Pyrolytic Graphite
GIC Graphite Intercalation Compound
GNR Graphene Nanoribbon
GO Graphene Oxide
GONR Graphene Oxide Nanoribbon
GOR Glucose Oxidation Reaction
vii
H2O2 Hydrogen Peroxide
H2SO4 Sulfuric acid
HAuCl4 Hydrogen Tetrachloroaurate (III) Tetrahydrate
HOPG Highly Ordered Pyrolytic Graphite
HOO Hydrogen peroxide anion
HOO
Hydroperoxy radical
K3[Fe(CN)6] Potassium Ferricyanide
KMnO4 Potassium Permanganate
LH Luminol monoanion
LH
Diazasemiquinone radical
LOD Limit of detection
LSV Linear Sweep Voltammetry
Luminol 3-amino-phthalhydrazide
MWCNT Multi-walled Carbon Nanotube
MWCNT_L Large Diameter of Multi-walled Carbon Nanotube (110-170 nm)
MWCNT_M Medium Diameter of Multi-walled Carbon Nanotube (40-70 nm)
MWCNT_S Small Diameter of Multi-walled Carbon Nanotube (3-20 nm)
N2 Nitrogen
NADH -nicotinamide adenine dinucleotide dehydrogenase
NHE Normal Hydrogen Electrode
O2 Oxygen
O2
Superoxide radical
OH Hydroxyl radical
OHads Hydroxyl adsorption
PBS Phosphate Buffer Solution
viii
PGO Pristine Graphite Oxide
PL Photoluminescence
PPy Polypyrrole
RE Reference Electrode
SEM Scanning Electron Microscopy
SPE Screen Printed Electrode
UA Uric acid
UV-Vis Ultraviolet Visible Spectrophotoscopy
WE Working Electrode
ix
NOMENCLATURE
Ep Peak-to-peak redox potential [V]
max Absorption peak [a.u.]
A Active surface area [cm2]
D0 Diffusion coefficient [cm2 s
-1]
Ip Peak current [A]
Ipeak Peak current density [mA cm-2
]
kobasal Electron transfer rate at basal plane [s
-1]
koedge Electron transfer rate at edge plane [s
-1]
v Scan rate [V s-1
]
Vox Oxidation peak [V]
1
CHAPTER 1
INTRODUCTION
1.1 Biosensors
Biosensors have been at the forefront of research due to their wide range of applications such
as in medical diagnosis, food inspections, environmental monitoring and many more [1].
Biosensors can be defined as simple analytical devices that combine the high sensitivity and
specificity of a biological molecule with the versatility of physical transducers to convert
biological responses into readable electronic signals [2]. A typical biosensor consists of two
parts, which are a biological sensing element and a physical transducer, as depicted in Fig.
1.1. The sensors recognition element enables the selective response to a particular analyte,
thus minimizing interference from other sample components [3]. The purpose of the
transducer is to convert a biochemical signal into an electronic signal [4]. There are various
types of transducer including piezoelectric, calorimetric, optical, electrochemical and thermal
transducer [2,4,5]. The appropriate transducer configuration depends significantly on the
nature of the biocatalyst system and the secondary products to be monitored [2]. In this
thesis, we focus on electrochemical transducers.
Figure 1.1: Basic building blocks of a biosensor [3].
2
1.1.1 Electrochemical biosensors
Electrochemical biosensors combine biological recognition element (enzymes, proteins,
antibodies, nucleic acids, etc) that selectively reacts with the target analytes and produces an
electrical signal whose intensity is proportional to the concentration of the analyte being
studied (Fig. 1.2 (A)) [3]. The electrochemical biosensor is part of an electrochemical cell,
which consists of three electrodes; working, reference and counter electrodes (Fig. 1.2 (B)).
The working electrode (WE) is where the reaction of interest occurs while the counter
electrode (CE, also known as auxiliary electrode) is used to close the current circuit in the
electrochemical cell, which is usually made of an inert material and does not participate in the
electrochemical reaction. The reference electrode (RE) is an electrode that has a stable and
well-known electrode potential, and is used as a point of reference in the electrochemical cell
for potential control and measurement [6]. The electrochemical reaction being monitored
usually generates a measurable current (amperometry), charge accumulation (potentiometry)
or resistance and reactance (impedance) [2-3]. The most common technique employed in
electrochemical sensing is amperometry, where a potential relative to the RE potential is
applied to WE in order to measure resultant current (Fig. 1.2 (C)) In amperometry, the
kinetics of the electron transfer reaction is driven by the applied potential, which affects the
diffusion-controlled current flowing across the electrode/solution interface. Consequently, the
current is obtained through electrolysis by means of an electrochemical reduction or
oxidation at WE, which is directly proportional to the bulk concentration of the analyte
present in the solution [6]. Therefore, electrochemical biosensors offer ease of fabrication and
sensitive detection of analytes, which covers a wide range of biomolecules [7]. The
electrochemical method is further improved by combination with the chemiluminescence
technique to achieve highly sensitive detection, as will be discussed in the next section.
3
Figure 1.2: (A) Principle of electrochemical biosensors. (B) An electrochemical cell. (C)
Output current based on oxidation-reduction (redox) reaction.
1.1.2 Electrochemiluminescence biosensors
Luminescence is the generation of light without heat. This phenomenon can occur through
several luminescent processes, namely photoluminescence (PL), chemiluminescence (CL)
and electrogenerated chemiluminescence (ECL), as depicted in Fig. 1.3 [8]. PL is a process in
which a substance absorbs photons (electromagnetic radiation) and then re-radiates photons,
which normally happens in solid materials [9]. CL is initiated by the mixing of necessary
reagents and is often controlled by the careful manipulation of fluid flow, while luminescence
in ECL is initiated and controlled by changing an electrode potential [8]. The advantages of
ECL method in comparison to CL are (i) the electrochemical reaction allows for the time and
position control of light-emitting reaction, (ii) it is highly selective due to the redox reactions
dependence on the applied potential, and (iii) it is non-destructive techniques [8].
4
Figure 1.3: Schematic of the general mechanisms of photoluminescence, chemiluminescence
and electrogenerated chemiluminescence [8].
Electrogenerated chemiluminescence (ECL) is produced from the excited state of an
ECL luminophore using electrochemical techniques, has been applied to the fields of
immunoassay, clinical sensing, and environmental monitoring owing to its high sensitivity
and extremely wide dynamic range [10]. ECL luminophores are mainly divided into organic
(e.g. Luminol) and inorganic (e.g. Ruthenium complex) compounds [11]. ECL is a means of
converting electrical energy into radiative energy that involves the production of reactive
intermediates from stable precursors at the surface of an electrode. These intermediates then
react under a variety of conditions to form excited states that emit light [9,11]. There are two
dominant pathways to generate ECL, namely the annihilation and co-reactant pathways [9-
11].
5
In the annihilation pathway, both oxidized and reduced species are produced on the
electrode surface by a potential step or sweep. These species then interact to produce a
ground state and an electronically excited state the latter of, which then relaxes by emission
(Eqs. 1-4) [10-11].
D e D
+ (oxidation at electrode) (1)
A + e A
(reduction at electrode) (2)
A
+ D+
A + D (excited-state formation) (3)
A A + hv (light emission) (4)
where A and D can be from the same species. In the co-reactant pathway, ECL is usually
generated by one directional potential scanning on the electrode in the presence of both the
luminophore and co-reactant (Eqs. 5-8). Common co-reactants used in ECL are the oxalate
ion (C2O42
) and tripropylamine (TPrA). A very interesting phenomenon in this system is that
the oxidation of the co-reactant leads to the generation of a strong reductant, instead of an
oxidant. ECL systems involving the generation of reductants by electrochemical oxidation are
referred to as oxidative-reductive ECL systems [9-10].
C2O42-
- e [C2O4
] CO2
+ CO2 (Co-reactant) (5)
D e D
+ (Luminophore)(6)
CO2
+ D+
D + CO2 (7)
D D + hv (8)
6
1.1.3 Nanomaterial-based electrochemical biosensors
The signal transduction and the general performance of electrochemical sensors are often
determined by the nanometer-scale surface architectures that connect the sensing element to
the biological sample [12]. One of the efforts is utilization of nanomaterials due to their
desirable properties. In particular, the ability to tailor the size and structure and hence the
properties of nanomaterials offers excellent prospects for designing novel sensing systems
and enhancing the performance of bioanalytical assays [13]. Nanomaterials can help address
some of the key issues in the development of all biosensors. Such issues include: design of
the biosensing interface so that the analyte selectively interacts with the biosensing surface;
achievement of efficient transduction of the biorecognition event; increases in the sensitivity
and selectivity of the biosensor; and improvement of response times in very sensitive systems
[14]. Electrochemical biosensors in combination with nanomaterials have become simple,
efficient tools to measure the concentration of analytes at low cost [7,13].
Nanomaterials, particularly carbon nanomaterials, have been at the forefront of
nanomaterial research for decades with the discovery of fullerenes, carbon nanotubes (CNTs),
and more recently, graphene [14]. Carbon nanomaterials have shown attractive properties for
electrochemical biosensing due to their high conductivity, large active surface area and ease
of functionalization [14-15].
Figure 1.4: Assembly of functionalized-single-walled CNT (SWCNT) electrically contacted
glucose oxidase electrode [16].
7
In addition, many kinds of nanoparticles, such as metal, oxide and semiconductor
nanoparticles have been used for constructing electrochemical sensors and biosensors. These
nanoparticles play different roles in different sensing systems. Among important functions
provided by nanoparticles include the immobilization of biomolecules, the catalysis of
electrochemical reactions, the enhancement of electron transfer between electrode surfaces
and proteins, labeling of biomolecules and even acting as reactants [17]. Catalysis is the most
important chemical application of metal nanoparticles, and has been studied extensively.
Transition metals, especially precious metals, have catalytic activity for many organic
reactions. In addition, metal nanoparticles can be used as heterogeneous and homogeneous
catalysts. The catalysis takes place on the active sites of the surface of metal nuclei (i.e. the
mechanism is similar to conventional heterogeneous catalysis).
Figure 1.5: Gold nanoparticle used as label in immunosensor [18].
1.2 Gold nanoparticle in electrochemical biosensor
Nanosized particles of noble metals, especially gold nanoparticles (AuNPs), have received
great interest due to their attractive electronic, optical, and thermal properties as well as
catalytic properties and potential applications in various fields [19]. Bulk gold (Au) is
chemically inert and is generally regarded as a poor catalyst. Interestingly, when Au is in the
form of very small particles with diameters below 10 nm and is deposited on metal oxides or
activated carbon, it becomes catalytically active, especially at low temperatures [20]. The
8
catalytic performance of Au is defined by three major factors; contact structure, support
selection and particle size [20-21]. The unique properties of AuNPs to provide a suitable
microenvironment for biomolecule immobilization while retaining their biological activity,
and to facilitate electron transfer between immobilized proteins and electrode surfaces, have
led to an intensive use of this nanomaterial for the construction of electrochemical biosensors
with enhanced analytical performance with respect to other biosensor designs [22].
Typically, AuNPs are prepared by chemical reduction of the corresponding transition
metal salts, hydrogen tetrachloroaurate (III) tetrahydrate (HAuCl4.4H2O), in the presence of a
stabilizer that binds to their surface to impart high stability and rich linking chemistry and to
provide the desired charge and solubility properties. After the breakthrough synthesis of
AuNPs reported by Schmid [23-24] and Brust et al. [25-26], a variety of methods has been
developed to prepare AuNPs. In our study, two types of syntheses were used to prepare
AuNPs; burst nucleation and seed growth methods for electrochemical- and
electrochemiluminescence-based biosensors, respectively. The burst nucleation method used
tert-butylamine-borane complex (TBAB) as a reducing agent as shown in Fig. 1.6. This
method required the utilization of hexane as a surfactant to avoid aggregation of AuNPs [27].
Figure 1.6: Burst nucleation method [27].
9
On the other hand, the seed growth method involved direct reduction of HAuCl4 in N-
(aminobutyl)-N-(ethylisoluminol) (ABEI) solution [28]. Through this method, AuNPs were
form with ABEI molecules existed on its surface through the covalent interaction between
gold and nitrogen atoms in their amino groups. Moreover, the carboxylic group in a molecule
of the oxidation product of ABEI results on electrostatic repulsion for the stabilization of
AuNPs (Fig. 1.7).
Figure 1.7: Seed growth method [28].
In chapter 3, the role of AuNPs in catalyzing non-enzymatic glucose oxidation
reaction was studied in neutral conditions. In alkaline conditions (Fig. 1.8), strong adsorption
of hydroxide anions on AuNPs enhanced the rate of deprotonation/dehydrogenation that
initiates the glucose oxidation reaction [29]. Regarding this, at neutral or low pH, the rate of
deprotonation/dehydrogenation is lower than at higher pH. However, it would be preferable
and allow for wider applications if commercialized products were designed to work at neutral
pH. To solve this problem, we focused on the development of functionalized carbon
nanomaterial which can act as an active functional supporting matrix that could promote the
dehydrogenation process and enhance the glucose oxidation reaction kinetics under neutral
10
conditions.
Figure 1.8: Schematic illustration of Au catalytic cycle for glucose oxidation reaction in
alkaline solution [29].
Luminol-H2O2 CL reaction has been widely applied for the detection of various
substances. It has been found that AuNPs could enhance the CL from the luminol-H2O2
system depending on their sizes [30]. The CL enhancement by AuNPs of the luminol-H2O2
system is proposed to originate from the particles catalytic activity, which facilitates the
radical generation and electron-transfer processes taking place on their surface [30]. Most
studies investigate ECL of luminol in solution on the surface of Au electrodes, leaving space
to explore ECL of luminol at the solid-liquid interface. In chapter 4, the seed growth method
produced isoluminol molecules coated on the surface of AuNPs, which were subsequently
drop cast on the working electrode surface. These isoluminol coated AuNP on modified
electrodes were subjected to cyclic voltammetry (CV) under various test conditions in order
to elucidate the ECL mechanism.
11
Figure 1.9: Mechanism for luminol-H2O2-gold colloid CL system [30].
Conjugation of AuNPs with other nanomaterials and biomolecules is an attractive
research area within nano-biotechnology [31]. In this context, carbon nanomaterials,
especially carbon nanotubes (CNTs), have attracted much interest in fundamental and applied
research due to their unique properties. Hybrid nanoparticle/nanotube materials have been
shown to possess interesting properties, which can be beneficial for the development of
electrochemical biosensors. Moreover, the incorporation of enzymes into the new composite
matrix enables the preparation of a mediator-less biosensor with a remarkably higher
sensitivity [32]. Hybrid composites can be prepared by selective attachment of AuNPs to
carbon nanotubes surfaces. This requires the prior of functionalization CNTs in order to
immobilize AuNPs. Based on the multiple functions of carbon nanomaterials in
electrochemical biosensors, we chose a recently discovered carbon material, graphene, as the
functional supporting matrix to improve the catalytic activity of AuNPs.
12
Figure 1.10: Schematic representation of the mechanism of nucleation of gold nanoparticles
at functionalized graphene surface [33].
1.3 Graphene for electrochemical applications
1.3.1 Graphene
Graphene is a single atomic layer of carbon atoms in an sp2 hexagonal bonding configuration.
It is the basic building block for other carbon materials, as illustrated in Fig. 1.11 [34]. Even
though graphene has been theoretically studied for decades, this material had never been
produced until 2003 by Andre Geim and Kostya Novoselov from University of Manchester
using the exfoliation method on graphite by scotch tape. Their groundbreaking experiments
regarding the 2D material graphene has won them the 2010 Nobel Prize in Physics.
Figure 1.11: Graphene is 2D building material for 0D fullerene, 1D nanotube and 3D graphite
[22]
13
Considerable research has been conducted on graphene since its first introduction by
Geim and Novoselov, in order to develop scalable synthesis methods [35-36] and improve
material properties [37-38] and functionalities [39-40]. Graphene can be synthesized mainly
by mechanical and chemical approaches as depicted in Fig. 1.12. Micromechanical
exfoliation has yielded small samples of graphene that are useful for fundamental study.
Although large-area films (up to ~1 cm2) made of single- to few-layer graphene have been
generated by chemical vapor deposition (CVD) growth on metal substrates, the uniform
growth of single-layer graphene is still a challenge [35]. Meanwhile, chemical synthesis
focuses on chemical exfoliation of graphite or cutting of carbon nanotubes to form graphene.
Exfoliation of graphite to graphite oxide through Hummers method has gained much interest
due to its simplicity; involves oxidation of graphite in the presence of strong acids and
oxidants. Subsequently, the oxygen-based functional groups on graphite oxide can be
removed by a reduction process.
Figure 1.12: Mechanical synthesis through (A) micromechanical exfoliation and (B) CVD
epitaxial growth. Chemical synthesis through (C) graphite oxidation and (D) MWCNT
unzipping.
14
1.3.2 Electrochemical performance of graphene
The properties and possible applications of graphene-based materials depend greatly on their
synthesis procedure (mechanically or chemically generated) [37]. An electrochemical
performance analysis of mechanically and chemically derived graphene demonstrates that
pristine graphene has a lower catalytic activity than graphene oxide or bulk graphite [41-42].
Recently, electron transfer in graphene has been demonstrated contributed by its peripheral
edge as opposed to its side [43]. This is because the former acts electrochemically akin to that
of edge plane- and the latter to that of basal plane-like sites/defects of highly ordered
pyrolytic graphite (HOPG). In HOPG, the electron transfer kinetics of the edge is
overwhelmingly dominant over that of the basal. Fig. 1.13 illustrates a schematic
representation of graphene (A) and graphite (B) indicating their edge and basal sites, where
the heterogeneous electron transfer rate of the former (ko
edge) is anomalously faster over that
of the latter (kobasal). The CVs in Fig. 1.9 depict that increasing the mass deposition of
graphene onto underlying edge-plane pyrolytic graphite (EPPG) electrode surface leads to a
decrease in the voltammetric peak height (Ip) as well as a significant decrease in the
electrochemical reversibility of the redox probe, as evident from the increasing peak-to-peak
redox potential (Ep). These results indicate that the addition of graphene onto the EPPG
electrode results in the reduction of the observed electron transfer kinetics. It is well
documented that the voltammetric response of graphitic electrodes depends on the proportion
of edge plane sites, where a low and high proportion result in slow and fast electron transfer,
respectively [41-42]. Consequently, the observed deterioration in the electrochemical
responses are likely due to the reduced proportion of available edge plane sites and an
increased basal plane contribution from the addition of graphene.
15
Figure 1.13: Effects of edge and basal planes on the electrochemical performance of (A)
pristine graphene and (B) graphite. Cyclic voltammetric profiles recorded utilizing 1 mM
potassium ferrocyanide (II) in 1 M KCl.
The majority of graphene used in electrochemistry is produced through the reduction
of graphene oxide (GO), which results in partially functionalized graphene sheets or
chemically reduced GO, since pristine graphene is inert and difficult to immobilize on
electrode surfaces [7]. Therefore, it is expected that reactive edges and structural defects or
impurities are beneficial for electrochemical activity [41-42]. Hence, we have chosen to focus
on graphene oxide nanoribbons (GONRs), which contain various oxygen-based functional
groups on both the basal plane and straight edges of the graphene strip, for use in
electrochemical sensors.
16
1.3.3 Graphene oxide nanoribbons
Graphene nanoribbons (GNRs) are strips of graphene with a high length-to-width ratio and
straight edges [44]. Experimentally, microscopic quantities of few-layer GNRs were made
available through several processes including microfabrication on graphite surfaces followed
by exfoliation, exfoliation of bulk graphite in the presence of surfactants or plasma etching of
multi-walled CNTs (MWCNTs) partially imbedded in a layer of protective polymer. Carbon
nanotubes are often described as rolled up graphene sheets; therefore, it would seem natural
to unroll them to obtain graphene [45]. Fig. 1.14 illustrates several methods utilizing
MWCNTs to yield GNRs.
Figure 1.14: Graphene nanoribbon synthesis method [45].
Among the aforementioned methods, the longitudinal unzipping of MWCNTs through
an oxidizing process, proposed by Kosynkin et al. [46], is extremely simple, efficient, and
potentially scalable for producing exfoliated graphene sheets. This unzipping method
involves suspending the MWCNTs in concentrated sulfuric acid (H2SO4) for 112 h followed
17
by a potassium permanganate (KMnO4) treatment. They determined that their method gained
a high degree of consecutive tube openings when high amounts of KMnO4 were used. Since
the unzipping process is oxidative, the resultant nanoribbons possess various types of
oxygen-based functional groups at their edges and on their graphitic surfaces, similar to GO.
Therefore, these nanoribbons are termed graphene oxide nanoribbons (GONRs). The
advantages of this method include the ability to control the number of graphene layers based
on the selected MWCNT diameter and the uniformity of the resultant nanoribbon structure.
Figure 1.15: Longitudinal unzipping of MWCNT [34].
Recent reports on the mechanism of GO formation from bulk graphite using a
modified Hummers method (Fig. 1.16) revealed that the first step was the conversion of
graphite to a sulfuric acidgraphite intercalation compound (H2SO4-GIC) followed by the
conversion of the GIC to the oxidized form of graphite (pristine graphite oxide, PGO) by
18
permanganate, and finally the formation of GO by the reaction of PGO with water [47]. The
formation of the H2SO4-GIC was described as beginning immediately upon exposure of the
graphite to the acid medium, while the conversion of GIC to PGO was slower depending on
the graphite source.
Figure 1.16: Schematic of conversion of bulk graphite to GO [47].
In comparison to graphite, MWCNTs as the starting material have a unique 2D
structure, with diameters ranging from 2 to 500 nm and lengths from 50 nm to a few mm. The
MWCNTs contain several concentric coaxial graphene cylinders with an interlayer spacing of
0.34 nm, similar to graphite [48]. However, studies have shown that this interlayer spacing
increases with a decreasing nanotube diameter as depicted in Fig. 1.17 [49]. Smaller nanotube
diameters have a higher interlayer spacing due to the large repulsive forces between adjacent
tubes, which result from their high curvature. In the case of intermediate and large-diameter
MWCNTs, the interlayer spacing varies based on the innermost tubes diameter with an
asymptotic limit of 0.344 nm [49]. Moreover, the turbostratic stacking of the constituent
19
shells in the MWCNT facilitate slipping of adjacent graphene shells, reducing the shear
modulus due to the weak van der Waals interactions [50]. Therefore, the degree of oxidation
and unzipping rates for nanotubes with different diameters are expected to vary.
Figure 1.17: Effect of MWCNT diameter in interlayer spacing and tubes properties [49].
Herein, in chapter 2 we investigate the effect of the suspension period in H2SO4 on
various diameters of MWCNTs using the longitudinal unzipping method to produce GONRs
for potential sensor applications. The resultant GONR products were characterized by
observing their surface morphology, analyzing their degree of oxidation, and disorder density
in comparison to their respective MWCNT precursors. Finally, the applicability of the
obtained GONRs as electrode materials was evaluated by the electrochemical detection of
hydrogen peroxide (H2O2) and -nicotinamide adenine dinucleotide dehydrogenase (NADH),
which are commonly used in evaluating biosensor systems.
20
1.4 Organization of thesis
This thesis starts with a brief introduction to biosensors and the materials used in this
research including gold nanoparticles (AuNPs) and graphene oxide nanoribbons (GONRs). In
chapter 2, the factors that affect the rate of unzipping and oxidation degree are discussed
thoroughly, specifically on multi-walled carbon nanotube (MWCNT) diameter and pre-
oxidation time. Subsequently, a report on the performance of the produced GONRs as
electrode materials in electrochemical biosensors is presented. In chapter 3, the tuning of
AuNP catalytic activity by GONR as a supporting matrix in a non-enzymatic glucose sensor
at neutral conditions is reported. Then, a thorough study on ECL of isoluminol-functionalized
AuNP on GONR and a proposed ECL mechanism are described in chapter 4. The potential
applications of isoluminol-functionalized AuNP with GONR on modified screen printed
electrodes in enzymatic biosensor are demonstrated. Finally, a summary on the findings and
future remarks are given in chapter 5.
1.5 References
[1] A.P.F. Turner, Biosensor: sense and sensibility, Chem. Soc. Rev. 42 (2013) 3184.
[2] C.R. Lowe, An introduction to the concepts and technology of biosensors, Biosensors 1
(1985) 3.
[3] N.J. Ronkainen, H.B. Halsall, W.R. Heineman, Electrochemical biosensor, Chem. Soc.
Rev. 39 (2010) 1747.
[4] J.E. Frew, H.A.O. Hill, Electrochemical biosensors, Anal. Chem. 59 (1987) 933.
[5] F.W. Scheller, R. Hintsche, D. Pfeiffer, F. Schubert, K. Riedel, R. Kindervater, Biosensors:
Fundamental, Applications and Trends, Sensors and Actuators B 4 (1991) 197.
[6] A.J. Bard, L.R. Faulkner, Electrochemical Methods Fundamentals and Applications, John
Wiley & Sons, Inc., 2nd
edn. (2001).
21
[7] K. Kerman, M. Saito, S. Yamamura, Y. Takamura, E. Tamiya, Nanomaterial-based
electrochemical biosensor for medical applications, Trends in Analytical Chemistry 27 (2008)
585.
[8] W. Miao, Electrogenerated chemiluminescence and its biorelated applications, Chem. Rev.
108 (2008) 2506.
[9] A.J. Bard, Electrogenerated Chemiluminescence, Marcel Dekker, Inc., 10th
edn. (2004).
[10] L. Hu, G. Xu, Applications and trends in electroclemiluminescence, Chem. Soc. Rev. 39
(2010) 3275.
[11] M.M. Richter, Electrochemiluminescence (ECL), Chem. Rev. 104 (2004) 3003.
[12] D. Grieshaber, R. MacKenzie, J. Vrs, E. Reimhult, Electrochemical Biosensors -
Sensor Principles and Architectures, Sensors 8 (2008) 1400.
[13] J. Wang, nanomaterial-based electrochemical biosensors, Analyst 130 (2005) 421.
[14] W. Yang, K.R. Ratinac, S.P. Ringer, F. Braet, Carbon nanomaterials in biosensors,
Angew. Chem. Int. Ed. 49 (2010) 2114.
[15] X. Yang, B. Feng X. He, F. Li, Y. Ding, J. Fei, Carbon nanomaterial based
electrochemical sensors for biogenic amines, Microchim Acta 180 (2013) 935.
[16] F. Patolsky, Y. Weizmann, I. Willner, Long-Range Electrical Contacting of Redox
Enzymes by SWCNT Connectors, Angew. Chem. Int. Ed. 43 (2004) 2113.
[17] X. Luo, A. Morrin, A.J. Killard, M.R. Smyth, Application of nanoparticles in
electrochemical sensors and biosensors, Electroanalysis 18 (2006) 319.
[18] K. Idegami, M. Chikae, K. Kerman, N. Nagatani, T. Yuhi, T. Endo, E. Tamiya, Gold
Nanoparticle-Based Redox Signal Enhancement for Sensitive Detection of Human Chorionic
Gonadotropin Hormone, Electroanalysis 20 (2008) 14.
[19] S. Guo, E. Wang, Synthesis and electrochemical applications of gold nanoparticles,
Anal. Chim. Acta 598 (2007) 191.
22
[20] M. Haruta, When gold is not noble: catalysis by nanoparticles, The Chemical Record 3
(2003) 75.
[21] A. Corma, H. Garcia, Supported gold nanoparticles as catalyst for organic reactions,
Chem. Soc. Rev. 37 (2008) 2096.
[22] J.M. Pingarrn, P. Yez-Sedeo, A. Gonzlez-Corts, Gold nanoparticle-based
electrochemical biosensors, Electrochimica Acta 53 (2008) 5848.
[23] G. Schmid, Large clusters and colloids, Chem. Rev. 92 (1992) 1709.
[24] G. Schmid, L.F. Chi, Metal clusters and colloids, Adv. Mater. 10 (1998) 515.
[25] M. Brust, M. Walker, D. Bethell, D.J. Schiffrin, R.J. Whyman, Synthesis of thiol-
derivatised gold nanoparticles in a two-phase LiquidLiquid system, J. Chem. Soc. Chem.
Commun. 7 (1994) 801.
[26] M. Brust, J. Fink, D. Bethell, D.J. Schiffrin, C.J. Kiely, Synthesis and reactions of
functionalised gold nanoparticles, J. Chem. Soc. Chem. Commun. 16 (1995) 1655.
[27] J. Naruse, L.Q. Hoa, Y. Sugano, T. Ikeuchi, H. Yoshikawa, M. Saito, E. Tamiya,
Development of biofuel cells based on gold nanoparticle decorated multi-walled carbon
nanotubes, Biosensors & Bioelectronics 30 (2011) 204.
[28] D. Tian, H. Zhang, Y. Chai, H. Cui, Synthesis of N-(aminobutyl)-N-(ethylisoluminol)
functionalized gold nanomaterials for chemiluminescent bio-probe, Chem. Commun. 47
(2011) 4959.
[29] T. Ishimoto, Y. Hamatake, H. Kazuno, T. Kishida, M. Koyama, Theoretical study of
support effect of Au catalyst for glucose oxidation of alkaline fuel cell anode, Applied Surface
Science 324 (2015) 76.
[30] Z.F. Zhang, H. Cui, C.Z. Lai and L.J. Liu, Gold nanoparticle-catalyzed luminol
chemiluminescence and its analytical application, Anal. Chem. 77 (2005) 3324.
[31] E. Katz, I. Willner, Integrated NanoparticleBiomolecule Hybrid Systems: Synthesis,
23
Properties, and Applications, Angew. Chem. Int. Ed. 43 (2004) 6042.
[32] J. Manso, M.L. Mena, P. Yez-Sedeo, J.M. Pingarrn, Electrochemical biosensors
based on colloidal goldcarbon nanotubes composite electrodes, J. Electroanal. Chem. 603
(2007) 1.
[33] G. Goncalves, P.A.A.P. Marques, C.M. Granadeiro, H.I.S. Nogueira, M.K. Singh, J.
Grcio, Surface modification of graphene nanosheet with gold nanoparticles: the role of
oxygen moieties at graphene surface on gold nucleation and growth, Chem. Mater. 21 (2009)
4796.
[34] A.K. Geim, K.S. Novoselov, The rise of graphene. Nat Mater 6 (2007) 183.
[35] S. Park, R.S. Ruoff, Chemical methods for the production of graphenes. Nat
Nanotechnol 4 (2009) 217.
[36] M.J. Allen, V.C. Tung, R.B. Kaner, Honeycomb Carbon: A review of graphene. Chem.
Rev. 110 (2010) 132.
[37] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V.
Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films. Science 306
(2004) 666.
[38] C.N.R. Rao, A.K. Sood, R. Voggu, K.S. Subrahmanyam, Some novel attributes of
graphene. J. Phys. Chem. Lett. 1 (2010) 572.
[39] T. Kuila, S. Bose, A.K. Mishra, P. Khanra, N.H. Kim, J.H. Lee, Chemical
functionalization of graphene and its applications. Prog. Mater. Sci. 57 (2012) 1061.
[40] D. Chen, H. Feng, J. Li, Graphene oxide: preparation, functionalization and
electrochemical applications. Chem. Rev. 112 (2012) 6027.
[41] D.A.C. Brownson, C.E. Banks. Graphene electrochemistry: fabricating amperometric
biosensor. Analyst 136 (2011) 2084.
[42] D.A.C Brownson, A.C. Lacombe, M. Gmez-Mingot, C.E. Banks. Graphene oxide gives
24
rise to unique and intriguing voltammetry. RCS Advances 2 (2012) 665.
[43] P.M. Hallam, C.E. Banks, Quantifying the electron transfer sites of graphene,
Electrochemistry Communication 13 (2011) 8.
[44] A.L. Higginbotham, D.V. Kosynkin, A. Sinitskii, Z. Sun, J.M. Tour, Lower-defect
graphene oxide nanoribbons from multiwalled carbon nanotubes. ACS Nano 4 (2010) 2059.
[45] M. Terrones, Nanotubes unzipped, Nature 458 (2009) 845.
[46] D.V. Kosynkin, A.L. Higginbotham, A. Sinitskii, J.R. Lomeda, A. Dimiev, B.K. Price,
J.M. Tour, Longitudinal unzipping of carbon nanotubes to form graphene nanoribbons.
Nature 458 (2009) 872.
[47] A.M. Dimiev, J.M. Tour, Mechanism of graphene oxide formation. ACS Nano 8 (2014)
3060.
[48] S. Iijima, Helical microtubules of graphitic carbon. Nature 354 (1991) 56.
[49] C.H. Kiang, M. Endo, P.M. Ajayan, G. Dresselhaus, M.S. Dresselhaus, Size effect in
carbon nanotubes. Phys. Rev. Lett. 81 (1998) 1869.
[50] X. Sun, C.H. Kiang, M. Endo, K. Takeuchi, T. Furuta, M.S. Dresselhaus, Stacking
characteristics of graphene shells in carbon nanotubes. Phys. Rev. B 54 (1996) 629.
25
CHAPTER 2
EFFECT OF MWCNT DIAMETER AND CHEMICAL OXIDATION TIME ON
FORMATION AND PERFORMANCE OF GRAPHENE OXIDE NANORIBBONS
FOR ELECTROCHEMICAL BIOSENSOR APPLICATIONS
2.1 Introduction
Graphene is a single atomic layer of carbon atoms in an sp2 hexagonal bonding configuration
that is the basic building block for other carbon materials [1]. Considerable research has been
conducted on graphene since its first introduction by Geim and Novoselov, in order to
develop scalable synthesis methods [2-3] and improve material properties [4-5] and
functionalities [6-7]. The properties and possible applications of graphene based materials
depend greatly on their synthesis procedure (mechanically or chemically generated) [6].
Both the mechanical exfoliation and the high-temperature growth techniques can produce
high quality graphene with low throughput, but at a high cost [2]. In contrast, the chemically
derived approach yields high throughput graphene with graphitic defects at a low cost [2]. An
electrochemical performance analysis of mechanically and chemically derived graphene
demonstrates that pristine graphene has a lower catalytic activity than graphene oxide or bulk
graphite [8-9]. Furthermore, it is expected that the reactive edges and structural defects or
impurities are beneficial for electrochemical activity [8-9]. Hence, we have chosen to focus
on graphene oxide nanoribbons (GONRs), which contain various oxygen-based functional
groups on both the basal plane and straight edges of the graphene strip, for use in
electrochemical sensors.
Recently, graphene nanoribbons (GNRs) have been produced by a chemical unzipping
of multiwalled carbon nanotubes (MWCNTs) [10]. The advantages of this method include the
26
ability to control the graphene layer number based on the selected MWCNT diameter and the
uniformity of the resultant nanoribbon structure. Several well-known techniques for
producing GNRs include treating MWCNT with oxidizing agents under acidic conditions
followed by a reduction process [11], etching by an argon plasma [12], exfoliation based on
the intercalation of alkali-metal atoms [13], and cleaving with catalytic metal nanoparticles
[14]. Among the aforementioned methods, the longitudinal unzipping of MWCNTs through
the oxidizing process, proposed by Kosynkin et al. [11], is extremely simple, efficient, and
potentially scalable for producing exfoliated graphene sheets. This unzipping method
involves suspending the MWCNTs in concentrated sulfuric acid (H2SO4) for 112 h followed
by a potassium permanganate (KMnO4) treatment. They determined that their method gained
a high degree of consecutive tube openings when high amounts of KMnO4 were used. Since
the unzipping process is oxidative, the resultant nanoribbons possess various types of
oxygen-based functional groups at their edges and on their graphitic surfaces, similar to
graphene oxide (GO). Therefore, these nanoribbons are termed graphene oxide nanoribbons
(GONRs). Furthermore, the GONRs are highly soluble in water without the assistance of
dispersing agents, and form stable colloids due to the electrostatic repulsion from the
carboxylic and hydroxyl groups present on the GONR surface [15]. With these interesting
properties, GONRs hold significant potential through a facile surface modification to
chemically functionalize them for use in electrochemical sensor applications.
In order to achieve fully unzipped MWCNTs on a large scale, several optimization
studies have been performed, starting with optimization of the quantity of oxidizing agent,
KMnO4 [11]. Further optimization of this method focused on varying the acid medium,
reaction temperature, and KMnO4 reaction time in order to form GONRs with fewer defects
[16]. Through their experiments, they confirmed that a sufficient amount of H2SO4 (about 90
vol%) was crucial for the complete formation and exfoliation of nanoribbons, while the
27
addition of a second acid (about 10 vol%) could enhance the GONR quality. The suspension
of the MWCNTs in concentrated H2SO4 prior to KMnO4 treatment is believed to allow
penetration and intercalation of the H2SO4, enhancing the oxidation process of the oxidizing
agents. Recent reports on the mechanism of graphene oxide formation from bulk graphite
using a modified Hummers method revealed that the first step was the conversion of graphite
to a sulfuric acidgraphite intercalation compound (H2SO4-GIC) followed by the conversion
of the GIC to the oxidized form of graphite (pristine graphite oxide, PGO) by permanganate,
and finally the formation of GO by the reaction of PGO with water [17]. The formation of the
H2SO4-GIC was described as beginning immediately upon exposure of the graphite to the
acid medium, while the conversion of GIC to PGO was slower depending on the graphite
source. In comparison to graphite, MWCNTs as the starting material have a unique 2D
structure, with diameters ranging from 2 to 500 nm and lengths from 50 nm to a few mm. The
MWCNTs contain several concentric coaxial graphene cylinders with an interlayer spacing of
0.34 nm, similar to graphite [18]. However, studies have shown that this interlayer spacing
increases with a decreasing nanotube diameter [19]. Smaller nanotube diameters have a
higher interlayer spacing due to the large repulsive forces between the adjacent tubes
resulting from their high curvature. In the case of intermediate and large-diameter MWCNTs,
the interlayer spacing varies based on the innermost tubes diameter with an asymptotic limit
of 0.344 nm [19]. Moreover, the turbostratic stacking of the constituent shells in the MWCNT
facilitate slipping of adjacent graphene shells, reducing the shear modulus due to the weak
van der Waals interactions [20]. Therefore, the degree of oxidation and unzipping rates for
nanotubes with different diameters are expected to vary.
Herein, we investigate the effect of the suspension period in H2SO4 on various
diameters of MWCNTs using the longitudinal unzipping method to produce GONRs for
potential sensor applications. To the best of our knowledge, this is the first study on the
28
unzipping rate of various MWCNTs diameters. The unzipping technique was applied to three
diameters of MWCNTs, representing small (320 nm), medium (4070 nm) and large (110
170 nm) diameter. These selected MWCNTs were then suspended in concentrated H2SO4 for
3, 6, or 12 h prior to the KMnO4 treatment. The resultant GONR products were characterized
by observing their surface morphology, analyzing the degree of oxidation, and disorder
density compared with their respective MWCNT starting materials. Finally, the applicability
of the obtained GONRs was evaluated by the electrochemical detection of hydrogen peroxide
(H2O2) and -nicotinamide adenine dinucleotide dehydrogenase (NADH), which are
commonly evaluated in biosensor systems.
2.2 Experimental Procedure
2.2.1 Materials and reagents
MWCNTs with a diameter of 320 nm (S) and 4070 nm (M), the concentrated H2SO4,
KMnO4, 30% hydrogen peroxide (H2O2), ethanol, ether, and potassium ferricyanide
(K3[Fe(CN)6]) were obtained from Wako Pure Chemical Industries, Ltd., Japan. MWCNTs
with a diameter of 110170 nm (L), potassium chloride (KCl), and carbon sheets (CS) were
purchased from Sigma-Aldrich, USA. The -nicotinamide adenine dinucleotide (NADH) was
obtained from Oriental Yeast Co. Ltd., Japan. The phosphate buffer solution (PBS, 0.1 mol
L1
, pH 7.0) was prepared from KH2PO4 and K2HPO4 from Sigma-Aldrich, USA. Ultrapure
water was obtained from a Barnstead Nanopure water purification system (18 M, Thermo
Scientific, USA) and was used in all experiments.
2.2.2 GONR synthesis and characterization
To synthesize the GONRs, the different types of MWCNTs were exposed separately for
longitudinal unzipping by suspending them in concentrated H2SO4 and then treating them
29
with KMnO4 [11]. Three sizes of commercial MWCNTs were used in this experiment; the
small size MWCNTs (MWCNT_S) have diameters between 320 nm, the medium size
(MWCNT_M) have diameters of 4070 nm, and the large size (MWCNT_L) have diameters
of 110170 nm. The MWCNTs were suspended in concentrated H2SO4 for 3, 6, or 12 h,
followed by treatment with 500 wt% KMnO4. The mixture was stirred at room temperature
for 1 h and then heated between 55 to 70 C for an additional 1 h. After all of the KMnO4 was
consumed, the mixture was quenched by pouring it onto ice containing a small amount of
H2O2. The solution was then filtered through a polytetrafluoroethylene (PTFE) membrane.
The remaining solid was washed with HCl and ethanol/ether alternately between filtrations.
The final product was dried in vacuo.
All of the GONR products were imaged by scanning electron microscopy (SEM)
using a Strata DB 235 microscope (FEI Co.) The GONR products degrees of oxidation were
examined by an attenuated total reflection Fourier Transform reflection infrared (ATR-FTIR)
spectroscopy (FT-720, Horiba, Japan). Meanwhile, Raman Spectroscopy (Raman-11,
Nanophoton Corp., Japan) was used to check the structure and topology of the MWCNTs and
GONRs. The degree of remaining -conjugation in the GONR after the oxidation process was
determined by a UV-Vis spectrophotometer (UV-2550, Shimadzu, Japan).
2.2.3 GONRs in biosensor applications
The electrochemical performances of GONR-based electrodes were evaluated with an
Autolab potentiostat/galvanostat PGSTAT12 (Metrohm/Eco Chemie, Netherlands). The
working electrode was fabricated by decorating carbon sheet (CS) substrates with the GONR
materials (GONR/CS) by the drop cast method. First, 0.5 mg of the GONRs were sonicated
and dispersed in 1 mL of ethanol. Then, the mixture was drop cast on the surface of a 1 cm2
cleaned carbon sheet and dried at 60 C for 1 h. To clean the surface of the GONR/CS, the
30
electrode was subjected to a thermal treatment at 400 C for 2 h. A three-electrode cell was
used for all electrochemical measurements, with a platinum coil as the counter electrode,
Ag/AgCl as the reference electrode (BAS Inc., Japan), and the prepared GONR/CS as the
working electrode.
2.3 Results and Discussion
2.3.1 Morphology inspection
0.5 mg/mL samples of each GONR products and MWCNT starting material were sonicated
thoroughly in ethanol before deposition on carbon sheets. Then, the fabricated samples were
heated at 400 C for 2 h before the SEM imaging. The resultant GONR surface morphologies
are summarized in Table 2.1. The lengthwise cutting and unraveling of the nanotube
sidewalls was observed on the as-prepared GONRs and compared to the MWCNT starting
materials. As shown in the figures of Table 1, the rate of unzipping to yield straight edge
nanoribbons was clearly affected by the MWCNT diameter and their suspension time in
concentrated H2SO4 prior to the permanganate introduction. Initially, GONRs obtained from
the MWCNT_S show only a minor degree of opening, concentrated either in the center or in
tears at the end of the nanotubes, regardless of the H2SO4 suspension time. Meanwhile, the
GONRs derived from the MWCNT_M gradually produced fully opened nanoribbons as the
H2SO4 suspension time was increased. On the other hand, GONRs derived from the
MWCNT_L formed huge exfoliated sheets, where the size was independent of H2SO4
exposure time. These results suggest that the MWCNT_S suffer from unzipping difficulty
compared to the MWCNT_M and MWCNT_L, and this may be due to high van der Waals
interactions from the adjacent coaxial graphene basal planes, especially for nanotubes with
radii below 7 nm [21]. Moreover, incomplete unzipping of the MWCNT_S was observed
even after increasing the reaction time in the concentrated H2SO4, indicating a low formation
31
of the GIC-H2SO4 intermediate, which then resulted in a low oxidation efficacy to break apart
the graphitic plane. Furthermore, the SEM images demonstrate that most of the unzipping
process takes place at the ends of the nanotubes rather than at the planar walls, suggesting
there is a greater reactivity in the existing defects at the MWCNT_S end caps [22].
Table 2.1. SEM images of GONR products.
Parameters Diameter of MWCNT
S M L
Aci
d T
reatm
ent
Tim
e (h
ou
rs)
0
MWCNT_S
320 nm
MWCNT_M
4070 nm MWCNT_L
110170 nm
3
Stacked tubes
Minor unzipped Stacked ribbons & tubes
Partially opened
Exfoliated ribbons
Fully opened
6
Stacked tubes
Minor unzipped
Stacked ribbons & tubes
Partially opened
Exfoliated ribbons
Fully opened
12
Stacked tubes
Minor unzipped
Stacked ribbons
Fully opened
Exfoliated ribbons
Fully opened
32
In contrast to the MWCNT_S, the GONRs produced from the MWCNT_M that
underwent less than 12 h of acid treatment had remaining unzipped inner tubes intertwined
with thick nanoribbon stacks. Meanwhile, the MWCNT_M that underwent more than 12 h of
acid treatment (GONR12M) were able to completely unzip all of the MWCNT layers, but
these still consisted of multi-stack graphene layers. The unzipped MWCNT_M have a
slightly curved morphology to their layers instead of flat nanoribbons. These results
confirmed that the unzipping process initiated from the outermost layer of the MWCNT, with
the long acid treatment time facilitating the complete opening of all MWCNT layers [11].
The irregular GONR surface morphology produced by the MWCNT_M may result from the
graphene pieces flaking off from wall defects during sonication [22]. Nevertheless, a low
degree of GONR exfoliation was observed even after a 12 h suspension time in H2SO4. This
result demonstrates a slow diffusion rate of the permanganate into the MWCNT_M, leading
to a slow disruption of the graphene lattice to increase the interlayer spacing [17].
Consequently, strong van der Waals forces prevent the GONR layers from undergoing a
chemical-mechanical breaking during sonication. In the case of the MWCNT_L, the resultant
GONRs are huge straight edged nanoribbons, even with the minimum exposure time in
concentrated H2SO4. These results clearly demonstrate a high rate of MWCNT_L unzipping
due to a greater number of defects at the end caps, coupled with the large curved surface area
that together serve as reactive sites for oxidation [22]. Moreover, the complete unzipping of
all MWCNT_L nanotube layers confirmed the easy penetration of H2SO4 to form the GIC-
H2SO4 intermediate, resulting in a faster diffusion rate of the oxidizing agent to oxidize the
nested graphene cylinders [17]. The SEM images also reveal that the GONRs produced from
MWCNT_L at a longer suspension period in the H2SO4 (GONR12L) tended to form a
fractured graphene sheet seam between the separated lattice fragments through oxygen atom
bridging [23]. These oxygen-zipped ridges produce a crumpling of the GONR sheet.
33
Furthermore, a long oxidation treatment with the strong acid also caused the GONR12L to
fracture into shorter segments.
In summary, our analysis revealed that the unzipping rate was dependent on the
diameter of the MWCNTs, as MWCNT_L > MWCNT_M > MWCNT_S. The effect of
increasing the MWCNT suspension time in concentrated H2SO4 was not apparent for the
MWCNT_S and MWCNT_L, but it significantly improved the unzipping rate of the
MWCNT_M. Other factors that might also affect the unzipping rate are number of nanotube
layers and innermost nanotube diameter. The different MWCNT diameters affect the
permanganate diffusion rate into the nanotube layers, which subsequently contributed to the
varied degree of oxidation degree. To detail the degree of oxidation as well as the structural
changes, FTIR measurements were taken of all samples with the results discussed in the
following section.
2.3.2 FTIR characterization
The oxidative unzipping of the MWCNTs is caused by oxygen atoms randomly attaching to
the honeycomb lattice, primarily at the defect or edge sites, forming an epoxy bridge that
joins two adjacent carbon atoms [24]. This three-membered epoxide is severely strained due
to the configuration changes that occur as the planar sp2-hybridized orientation switches to a
distorted sp3-hybridized geometry. Therefore, the cooperative alignment of the epoxy groups
induces sufficient tension in the underlying lattice to break the native carbon bonds [24]. The
opening sites of the nanotubes are expected to be decorated with ketones that can then be
converted to carboxylic groups [11]. On the other hand, MWCNTs with different diameters,
having different numbers of layers, were predicted to show a diverse degree of oxidation, as
the oxidation reaction rate of the MWCNTs innermost layers is considerably slower than that
of the outermost layers [25]. Furthermore, the variation in the oxidation rate between the
34
layers may affect the GONR sheet exfoliation due to low intercalation of oxygen functional
groups, leading to a small spacing displacement that is insufficient to overcome the van der
Waals forces.
In order to examine the degree of oxidation for the GONR derived from MWCNTs
with different diameters and varied acid treatment time, we prepared a thin layer of each
material by dropping a 0.5 mg/mL GONR or MWCNT precursor solution in ethanol on the
ATRs diamond surface and allowing it to dry. Before the catalyst deposition, the control
sample was analyzed first in background mode then sample mode. The background was
subtracted from the sample mode signal allowing us to distinguish the surrounding air peaks
(Fig. 2.1 (A)) at ~2300 cm-1
that attributed to the asymmetric vibration stretch of CO2, while
the noisy signals in the regions 34004000 cm-1
and 13001900 cm-1
were derived from
water vapor [26-27]. In addition, all GONR spectra (Fig. 2.1 (B-D)) showed the common
hydroxyl band at ~3420 cm-1
(i) for the COO-H/O-H stretching [25-28] in addition to three
strong carbonyl bands at ~1700 cm-1
(ii) for C=O stretching [25-28], ~1564 cm-1
(iii) for
COO- asymmetric stretching [26], and ~1209 cm-1
(iv) for C-O stretching [25,28]. In
comparison, all MWCNT spectra (Fig. 2.1 (A)) displayed no common hydroxyl (i) or
carbonyl (ii-iv) peaks, indicating that all GONR products had undergone the oxidation
process. The GONR obtained from the MWCNT_S had the lowest hydroxyl band (i) intensity
with a relatively high carbonyl band (ii-iv) intensity, whereas the GONR derived from the
MWCNT_M showed an incremental increase in both the carbonyl (ii-iv) and hydroxyl (i)
peak intensities compared to the GONR from the MWCNT_S. Meanwhile, the GONR
produced from the MWCNT_L had a nearly equivalent carbonyl peak intensity (ii-iv) to that
of the other GONRS, but with the highest hydroxyl peak intensity (i). These results suggest
that the MWCNT_S suffered from a low degree of oxidation, leading to partial or minimally
unzipped MWCNTs, and therefore the hydroxyl groups were expected to appear at the
35
nanotube ends torn edges, as imaged by the SEM in Table 2.1. A higher degree of oxidation
was observed in the case of the MWCNT_M, indicated by the appearance of ketone groups,
which produce fully unzipped nanotubes, but with a low intercalation of hydroxyl groups on
the graphitic surface, and therefore a low exfoliation rate between the MWCNT layers. The
GONR produced by the MWCNT_L had the highest degree of oxidation, a completed
oxidation process that resulted in a high amount of carboxylic groups due to ketone
protonation, as observed by the strong hydroxyl band intensity [11].
Figure 2.1: ATR-FTIR spectra of (A) MWCNTs with different diameters, and GONR derived
from (B) MWCNT_S, (C) MWCNT_M, and (D) MWCNT_L at varied suspension times in
concentrated H2SO4.
36
With regard to the effect of the acid treatment time, the degree of oxidation for the
produced GONRs increased with increasing suspension time in concentrated H2SO4 for all
types of MWCNTs. The FTIR spectra of the GONRs synthesized from the MWCNT_S (Fig.
1B) showed a slight increase in the hydroxyl band (i) as the suspension time in H2SO4 was
increased from 3 to 12 h. However, a strong carbonyl band (iiiii) was observed for a short
suspension period in H2SO4, and the C-O moiety band (iv) had the strongest absorbance
intensity at the 3 and 12 h of acid treatment times. These results imply that different acid
treatments periods contribute to different amounts of the H2SO4-GIC intercalate that react
strongly with the large phenolic groups produced under strong KMnO4 oxidation in H2SO4.
As the oxidation process proceeds, the majority of these phenolic groups likely condense to
form C-O-C ether linkages or epoxy bridges (iv) in order to rupture the graphitic lattice. A
small portion of the phenolic groups at the edges or at defect sites were oxidized to ketone or
quinone groups (iiiii), and subsequently, only a few ketone groups are converted to
carboxylic groups, as referred to in (iiii). Since there is a low intensity of the O-H band (i),
we can conclude that band (iv) mainly contributed to epoxy bridges that failed to crack the
graphitic lattice, and carbonyl groups (iiiii) primarily decorated the edges of the nanotubes
end caps. On the other hand, the GONR produced from the MWCNT_M (Fig. 1C) shows a
strong absorbance intensity of bands (iiv) as the suspension period in concentrated H2SO4
was increased, especially the GONR12M sample. The significant increase in bands (i) and
(iv) indicates the presence of hydrated surface oxides from O-H deformation and C-O
stretching that may be attributed to phenols, hydroquinones, and aromatic carboxylic groups.
Carbonyl bands (iiiii) of the GONR synthesized after 3 and 6 h of exposure in H2SO4
(GONR3M and GONR6M) have nearly identical intensities while band (iv) shows a high
variation in intensity between acid treatment times. The GONR3M and GONR6M possess
low band (i) and high band (iv), indicating that C-O-C bridges were formed by an incomplete
37
unzipping process. In addition, the oxidative unzipping of MWCNT_L (Fig. 1D) produced
GONR with nearly equivalent absorbance intensities for all bands (iiv) depending on the
acid treatment time. The intensities of the hydroxyl, carbonyl, and ether bands increased
proportionally with longer acid treatment time, suggesting that the MWCNT_L has the
highest rate of oxidation and unzipping to form the GONR.
In conclusion, the pristine MWCNT properties such as diameter can have a significant
effect on the degree of oxidation, and therefore the unzipping rate. Moreover, long acid
treatment times increase the degree of oxidation and hydrophilicity of the GONR product. A
high hydrophilicity is vital for promoting exfoliation, particularly for creating fully unzipped
MWCNT. These results reveal that the order of increased degree of oxidation is MWCNT_L
> MWCNT_M > MWCNT_S and 12 h > 6 h > 3 h. The oxidative unzipping process
introduces of defects and disorder on the graphitic lattice, as investigated in the following
sections.
2.3.3 Raman analysis
Raman spectroscopy has been widely used to characterize carbonaceous materials and
evaluate the degree of disorder in their structure [30]. The relative degree of structural defects
and disorder in graphene is usually evaluated by analyzing the intensity ratio between the
disorder-induced D band and the sp2-reduced G band (ID/IG) [31]. To investigate the defects
and disorder of the prepared GONRs compared to their pristine MWCNT precursors, a few
drops of 0.5 mg/mL individual solutions of either the GONRs or MWCNTs in ethanol were
cast onto quartz micro-glasses. After drying, the sample was observed using the 514 nm
wavelength at room temperature. Fig. 2.2 (A) displays the Raman spectrum of pristine
MWCNT_L, showing the obvious G band (1565 cm-1
) resulting from the MWCNTs highly
crystalline structure and a minimal D band (1331 cm-1
) of amorphous or disordered carbon
38
[30-31]. In contrast, the GONR materials formed from the MWCNT_L show a broadened G
band after the oxidation process along with the appearance of the D band. This identifies a
reduction of the in-plane sp2 domain size and an increasing level of disorder in the GONR
due to the oxidation process [11]. Furthermore, this result correlates with the FTIR data in
Fig. 2.1 (D), where the oxidative unzipping of MWCNT_Ls by various acid treatment times
produced oxygen-based functional groups along the edges and basal planes, reducing the
honeycomb lattice structure.
Figure 2.2: (A) Raman spectra of the GONRs and their MWCNT_L precursor with various
suspension times in H2SO4. (B) The D to G band peak intensity ratio (ID/IG) with error bars
indicating the standard deviation of six ID/IG ratios across each sample.
The chemically modified MWCNTs have higher ID/IG ratios than pristine MWCNTs
as depicted in Fig. 2.2 (B) due to the high degree of induced defects and disorder. Among the
pristine MWCNTs, the MWCNT_L had the lowest ID/IG ratio compared to MWCNT_M and
MWCNT_S because of graphitization, which results in a polygonal cross section with low
structural defects [19]. Interestingly, the average ID/IG ratios of the GONR derived from the
different MWCNT diameters, following an order of MWCNT_S > MWCNT_M >
MWCNT_L, inferring an identical trend for the degree of defects and disorder in these
products. The degree of defects in the GONRs produced from the MWCNT_L was about
39
176% higher than their starting material, with the ID/IG ratio of the GONR3L being higher
than that of GONR6L and GONR12L. These results indicate that the MWCNT_L is easily
oxidized even after a very short exposure time to acid, and it has few original defects that act
as reactive sites in the oxidation process. Since the MWCNT_L has a much larger diameter,
the resultant GONRs also have larger remaining sp2 domains than the disorder domains,
which consequently contributes to their low ID/IG ratios compared to GONRs derived from
the MWCNT_M and MWCNT_S. In contrast, the MWCNT_S precursor produced 19% of
structural defects on the GONR, and increased the disorder level along with increased acid
treatment time. These results confirm that the smaller diameter nanotubes are less reactive
and result in a low oxidation rate [25]; hence, the longer suspension time in H2SO4 provides
more intercalant species and defects to promote the oxidation-based unzipping process. In the
case of the GONRs synthesized from the MWCNT_M, the degree of defects was about 27%,
with small variations in the ID/IG ratio across the varied concentrated H2SO4 suspension times.
This result is in agreement with the FTIR data in Fig. 1B, where the GONR12M and
GONR3M demonstrated a high degree of oxidation by the presence of hydroxyl and epoxy
bridges, respectively, which also led to high disorder levels. As the GONR conductivity is
directly affected by structural defects and disorder, the remaining amount of -conjugation in
the GONRs was investigated and discussed in the following section.
2.3.4 Investigation on the degree of conjugation
The characterization of the GONR by FTIR revealed that the MWCNTs underwent the
oxidation process while the Raman spectra showed that the GONRs had increased disorder
and oxygen-based functional group defects. Thus, the level of -conjugation remaining in the
GONR can be determined by the max of its absorption when dispersed in DI water using a
UV-Vis spectrophotometer. The absorption peak, max of the GONRs produced from the
40
MWCNT_L with various acid treatment times (Fig. 3A) and of GONRs synthesized from
various MWCNT diameters at 6 h of suspension in concentrated H2SO4 (Fig. 3B) appeared
around 240260 nm. A shoulder around ~300 nm can be observed and attributed to the n-*
transition of the carbonyl groups. It is known that highly conjugated graphene-like materials
have an max at ~275 nm while, materials with a disrupted -network and greater number of
sp3 carbons have a blue shifted max (~235 nm) [16, 24]. Based on Fig. 3A, the max value
shifts in the order of GONR3L > GONR12L > GONR6L, suggesting a high degree of
conjugation in the GONR6L. Both the UV-Vis and Raman results imply that the oxidative
unzipping process disrupts the crystalline MWCNT_L conjugation; however, the majority of
-conjugation can be maintained due to the vast size of the exfoliated GONR. In comparison,
the max of GONR synthesized from different MWCNT diameters shows a shift of GONR6L
> GONR6M > GONR6S (Fig. 3B). This result is in agreement with the degree of oxidation
revealed by FTIR and further elucidated the degree of disorder in the Raman spectra,
indicating that the GONR produced from the MWCNT_M and MWCNT_S were unopened
and stacked with low hydrophilicity. At this point, we have identified that the unzipping
process for various diameter MWCNT for different suspension times in concentrated H2SO4
produced GONRs with different degrees of oxidation, disorder, and -conjugation network.
Moreover, GONRs with a disrupted -conjugation network and oxygenated functional groups
have a poor conductivity [11]. Nevertheless, the defects and functionalization of the
graphene-based electrode has proven beneficial in electrochemical sensor applications [8].
The performance of GONR-based electrochemical sensors is demonstrated in the following
sections.
41
Figure 2.3: (A) UV-Vis absorption spectra of GONRs produced from MWCNT_L with
varied suspension times in H2SO4. (B) UV-Vis of GONRs synthesized from varied diameters
of MWCNTs with 6 h of H2SO4 treatment time.
2.3.5 Electrochemical performance and sensor applications
Several researchers have studied the relationship between surface structure and
electrochemical phenomena, especially for carbon electrodes. In general, the electrode
reactivity is influenced by the carbon materials microstructure [8], surface functional groups
[32], and surface defects [33]. Herein, the produced GONRs were explored as the electrode
materials in electrochemical sensors. First, a GONR-based electrode was tested in a
ferro/ferricyanide redox probe to provide information on the electron transfer kinetics, based
on the peak-to-peak redox potential (Ep) between the reversible redox peaks, peak current
density (Ip), and calculated active surface area (A). Then, the performance of the GONR-
based electrodes towards the electro-oxidation of H2O2 and NADH were investigated using
cyclic voltammetry (CV). H2O2 and NADH are essential mediators in enzymatic reactions,
and therefore, the sensitive detection of these components is important for developing
oxidase- and dehydrogenase-based electrochemical biosensors [34].
42
Figure 2.4: Extracted peak-to-peak redox separation (Ep) of GONR- or MWCNT-modified
CS. The electrolyte was 10 mM K3[Fe(CN)6] in 1 M KCl and the scan rate was 100 mV/s.
Fig. 2.4 (A) shows that CS modified with pristine MWCNTs have increased Ep
values, indicating slow electron transfer kinetics. This is due to the hydrophobic
characteristics of the MWCNTs, which hinder electrolyte diffusion. Furthermore, the Ep
values of the different MWCNT diameters varied as follows, MWCNT_M > MWCNT_L >
MWCNT_S. The MWCNT_L had faster electron transfer kinetics than the MWCNT_M,
likely due to the larger nanotubes providing a large surface area and easy electrolyte
penetration for chemical reactions. Moreover, the MWCNT_L is less hydrophobic than
MWCNT_M and MWCNT_S, as it produces a homogeneous layer on the CS electrode (Fig.
2.5). In contrast, the MWCNT_S has very fine tubes that reduce the CS electrodes surface
roughness, making it easier for the electrolyte to make contact compared to the MWCNT_M.
43
Figure 2.5: Electrode surface of bare carbon sheet (CS) and various MWCNT diameter
precursors modified CS.
Fig. 2.4 (B) displays decreasing of Ep values proportional to increasing in acid
treatment time for the GONRs yielded from the MWCNT_L, indicative of a fast electron
transfer at a higher degree of oxidation. Therefore, the Ip and A values also increase, as
depicted in Fig. 2.6 (B) and 2.7 (B), respectively. Since the Fe(CN)63-/4-
is insensitive to
surface oxide coverage [9, 32], the enhanced in conductivity and catalytic performance
mainly resulted from the huge active surface area and surface defects on the exfoliated
GONR, which are attractive as electrochemical reaction sites [31, 33-34]. The Ep, Ip, and A
values of the GONRs synthesized from the MWCNT_M were identical to those of the
GONRs formed from the MWCNT_L. Their electron transfer kinetics increased significantly
with longer acid treatment times, which may be due to the increased GONR hydrophilicity
enhancing the electrolyte contact with the electrodes surface. However, GONRs produced
from the MWCNT_S show increased Ep values for longer suspension times in H2SO4 (Fig.
2.4 (D)). Nonetheless, no changes were observed on their Ip and A values, as presented in Fig.
2.6 (D) and 2.7 (D), respectively, which indicates slow electron transfer kinetics and poor
electro-catalytic activity due to a low active surface area. This result agrees with the SEM
images in Table 2.1 and the FTIR spectra in Fig. 2.1 (D) that reveal minimal unzipping and a
low degree of oxidation for the MWCNT_S regardless of acid treatment times.
44
Figure 2.6: Extracted peak current density (Ip) of GONR- or MWCNT-modified CS. The
electrolyte was 10 mM K3[Fe(CN)6] in 1 M KCl and the scan rate was 100 mV/s.
The active surface area was calculated using Randle-Sevcik equation:
= 268600 3/20
1/201/2
where Ip, n, D0, C0 and v represent redox peak current (A), number of electrons transferred in
the redox event, diffusion coefficient (cm2 s
-1), concentration of redox species (mol cm
-3) and
scan rate (V s-1
) [19]. The D0 of 10 mM K3[Fe(CN)6] in 1 M KCl for carbon sheet (CS)
electrode is 7.33 105 cm2 s-1.
45
Figure 2.7: Calculated active surface area (A) of GONR- or MWCNT-modified CS. The
electrolyte was 10 mM K3[Fe(CN)6] in 1 M KCl and the scan rate was 100 mV/s.
Fig. 2.8 compares the background-subtracted CVs of the GONR-based electrodes in
the presence of H2O2. The onset potentials for oxidation/reduction currents for the CS,
GONR12S, GONR12M, and GONR12L began at 0.63/-0.18 V, 0.56/-0.11 V, 0.52/0.09 V,
and 0.50/0.12 V, respectively (Fig. 2.8 (A)). This result shows that the bare CS had a poor
catalytic activity, and modifying it with the GONR materials improved the electro-catalytic
performance towards the H2O2 redox reactions. Moreover, the high degree of oxidation and
edge-plane defects in the GONR12L may have contributed to its superior catalytic activity
compared to the GONR12M and GONR12S. Fig. 2.8 (B) displays the onset potentials for
oxidation/reduction currents for the MWCNT_L, GONR3L, GONR6L, and GONR12L,
which began at 0.65/-0.15 V, 0.56/0.07 V, 0.53/0.10 V, and 0.50/0.12 V, respectively. The
pristine MWCNT_L-modified CS electrode exhibited a low electro-catalytic activity
compared to the GONR-based electrode, indicating that the oxygen-based functional groups
46
and edge-plane defect sites are required to catalyze the H2O2 redox reaction [8, 34]. In
addition, this result also confirmed that the GONRs produced from longer acid treatment
times demonstrated a higher degree of oxidation, producing a more active catalytic material
for H2O2 detection.
Figure 2.8: (A) CVs of GONR-based electrodes from various MWCNT diameters precursors
after 12 h of H2SO4 treatment time. (B) CVs of GONR-based electrodes produced from the
MWCNT_L precursor after various suspension periods in H2SO4. The electrolyte is 1 mM
H2O2 in 0.1 M PBS (pH 7.0) with a scan rate of 50 mV/s.
Fig. 2.9 (A) shows that the oxidation of NADH occurs at 0.38 V for a CS. Modifying
the CS with a GONR material maintained this oxidation peak potential, but resulted in an
increased peak current density. The CVs show that the GONR12L has the highest peak
current density, indicating enhanced catalytic activity due to the exfoliated GONRs providing
a high surface reactivity and edge-plane defect density [34]. This phenomenon may also be
due to an increased effective surface area, manifested by an increase in the background
charging current [35] and verified by the calculation in Fig. 2.7. Furthermore, Fig. 2.9 (B)
suggests that the pristine MWCNTs have a lower electro-catalytic activity than the GONR
materials, indicating that the oxygen functional groups catalyzed the NADH oxidation
47
reaction [35]. Consequently, the GONR synthesized at longer suspension times in H2SO4
gave a higher electro-catalytic performance due to a high degree of oxidation and larger
active surface area.
Figure 2.9: (A) CVs of GONR-based electrodes from various MWCNT diameters precursors
after 12 h of H2SO4 treatment time. (B) CVs of GONR-based electrodes produced from the
MWCNT_L precursor after various suspension periods in H2SO4. The electrolyte is 1 mM
NADH in 0.1 M PBS (pH 7.0) with a scan rate of 50 mV/s.
In conclusion, the electrochemical performance of the GONR-based electrode is
significantly affected by its surface morphology and degree of oxidation, which vary through
the oxidative unzipping process of various MWCNT diameters in different acid treatment
times. A large proportion of oxygenated species present on the GONR can be beneficially
when used in electrochemical sensors where oxygenated electro-catalytic reactions are
employed. Based on these results, a higher performance of the electrodes is observed with the
modification of the GONRs, following the precursors of MWCNT_L > MWCNT_M >
MWCNT_S at 12 > 6 > 3 h of acid treatment times.
48
2.4 Conclusions
In summary, we have studied for the first time, the effects of MWCNT diameter and H2SO4
pre-treatment time towards GONR production by the oxidative unzipping method. SEM
images reveal that large-diameter MWCNTs have a faster rate of unzipping than smaller
MWCNT diameters. Moreover, only the medium-diameter MWCNTs depend on the acid
treatment period to produce fully unzipped nanoribbons. FTIR spectra disclose the degree of
oxidation, which have increasing peak intensities correlated to increase acid treatment times
and MWCNT diameters. This indicates that the large-diameter MWCNTs have highly
reactive surfaces that easily interact with oxygen atoms, subsequently cracking the graphitic
lattice. Thus, the degree of oxidation is correlated with the ID/IG ratio in Raman spectra,
showing 176%, 27%, and 19% increments in their degree of disorder for the GONR products
compared to MWCNT_L, MWCNT_M, and MWCNT_S, respectively, while a low ID/IG
ratio variation was observed for the different acid treatment times. Interestingly, the
maximum absorbance peak, max, shows that the highly oxidized GONR synthesized from the
MWCNT_L maintained a large amount of its -conjugated structure. Applying the GONRs as
electrode materials in electrochemical sensors demonstrated that they exhibited superior
electro-catalytic activity compared to their MWCNT precursors owing to the highly active
surface areas with oxygen-based functional groups that facilitate catalyzing redox reactions.
These enhance sensor performance, which in turn has been shown as being dependent on the
size of the original MWCNTs and treatment time in concentrated H2SO4. Further work on
improving the GONRs electro-catalytic activity by functionalization has commenced in our
laboratory, paving the way for future development of GONR-based electrochemical
biosensors.
49
2.5 References
[1] A.K. Geim, K.S. Novoselov, The rise of graphene. Nat Mater 6 (2007) 183.
[2] S. Park, R.S. Ruoff, Chemical methods for the production of graphenes. Nat Nanotechnol
4 (2009) 217.
[3] M.J. Allen, V.C. Tung, R.B. Kaner, Honeycomb Carbon: A review of graphene. Chem.
Rev. 110 (2010) 132.
[4] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V.
Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films. Science 306
(2004) 666.
[5] C.N.R. Rao, A.K. Sood, R. Voggu, K.S. Subrahmanyam, Some novel attributes of
graphene. J. Phys. Chem. Lett. 1 (2010) 572.
[6] T. Kuila, S. Bose, A.K. Mishra, P. Khanra, N.H. Kim, J.H. Lee, Chemical
functionalization of graphene and its applications. Prog. Mater. Sci. 57 (2012) 1061.
[7] D. Chen, H. Feng, J. Li, Graphene oxide: preparation, functionalization and
electrochemical applications. Chem. Rev. 112 (2012) 6027.
[8] D.A.C. Brownson, C.E. Banks. Graphene electrochemistry: fabricating amperometric
biosensor. Analyst 136 (2011) 2084.
[9] D.A.C Brownson, A.C. Lacombe, M. Gmez-Mingot, C.E. Banks. Graphene oxide gives
rise to unique and intriguing voltammetry. RCS Advances 2 (2012) 665.
[10] M. Terrones, Nanotubes unzipped. Nature 458 (2009) 845.
[11] D.V. Kosynkin, A.L. Higginbotham, A. Sinitskii, J.R. Lomeda, A. Dimiev, B.K. Price,
J.M. Tour, Longitudinal unzipping of carbon nanotubes to form graphene nanoribbons.
Nature 458 (2009) 872.
[12] L. Jiao, X. Wang, G. Diankov, H. Wang, H. Dai, Facile synthesis of high-quality
graphene nanoribbons. Nat Nanotechnol 5 (2010) 321.
50
[13] A.G. Cano-Mrquez, F.J. Rodrguez-Macas, J. Campos-Delgado, C.G. Espinosa-
Gonzlez, F. Tristn-Lpez, D. Ramrez-Gonzlez, D.A. Cullen, D.J. Smith, M. Terrones, Y.I.
Vega-Cant, Ex-MWNTs: Graphene sheets and ribbons produced by lithium intercalation and
exfoliation of carbon nanotubes. Nano Lett 9 (2009) 1527.
[14] Ci L, Zu X, Wang L, Gao W, Ding F, Kelly KF, et al. Controlled nanocutting of
graphene. Nano Res 1 (2008) 116.
[15] A.M. Dimiev, L.B. Alemany, J.M. Tour, Graphene oxide. Origin of acidity, its instability
in water and new dynamic structural model. ACS Nano 7 (2013) 576.
[16] A.L. Higginbotham, D.V. Kosynkin, A. Sinitskii, Z. Sun, J.M. Tour, Lower-defect
graphene oxide nanoribbons from multiwalled carbon nanotubes. ACS Nano 4 (2010) 2059.
[17] A.M. Dimiev, J.M. Tour, Mechanism of graphene oxide formation. ACS Nano 8 (2014)
3060.
[18] S. Iijima, Helical microtubules of graphitic carbon. Nature 354 (1991) 56.
[19] C.H. Kiang, M. Endo, P.M. Ajayan, G. Dresselhaus, M.S. Dresselhaus, Size effect in
carbon nanotubes. Phys. Rev. Lett. 81 (1998) 1869.
[20] X. Sun, C.H. Kiang, M. Endo, K. Takeuchi, T. Furuta, M.S. Dresselhaus, Stacking
characteristics of graphene shells in carbon nanotubes. Phys. Rev. B 54 (1996) 629.
[21] X.Q. He, S. Kitipornchai, C.M. Wang, K.M. Liew, Modeling of van der Waals force for
infinitesimal deformation of multi-walled carbon nanotubes treated as cylindrical shells. Int.
J. Solid Struct. 42 (2005) 6032.
[22] S.R. Dhakate, N. Chauhan, S. Sharma, R.B. Mathur, The production of multi-layer
graphene nanoribbons from thermally reduced unzipped multi-walled carbon nanotubes.
Carbon 49 (2011) 4170.
[23] P.M. Ajayan, B.I. Yakobson, Oxygen breaks into carbon world. Nature 441 (2006) 818.